CST-005337

23
4 7 2552

Simulation of Heat Transfer in a Computer Notebook
case to optimize the locations of Fan
1, 1, 1, 1 1*

1

. . 47000
* : : (6642) 725033-3562, : (6642) 725034
E-mail: cutieatom@gmail.com

(computer notebook)

(computational fluid dynamics, CFD)

(CPU), hard disk (finite volume
method) k
k realizable standard wall function


: , ,

Abstract
Most air cooling within a computer notebook flows are turbulent flow which is more complicated than
laminar flow Therefore, turbulence model is important for computational fluid dynamics (CFD) study of this
type of flow. This research therefore aims to simulate the central processing unit (CPU), hard disk and
memory cards. The finite volume method using k turbulence model is used in the current work. From
the calculation, the realizable k turbulence model calculated with the non-equilibrium wall function gives
the closest result to the experimental data compared with the other RANS turbulence models.
Keywords: Heat Transfer, Computer Notebook, Computational Fluid Dynamics

1. (computer notebook)


- (computer desktop)
 CST-005337



2.

3
(continuity equation) (momentum
equation) (energy equation)


standard, RNG,
realizable k

2.1
(Reynolds-averaged Navier-Stokes turbulence
model, RANS)
2000 J.S.Chaing et.al. [1]
2 (continuity equation)


2549
( ui ) = 0 (1)
xi
[2] FLUENT
(momentum equation)

ui u j

ui
t
+ ui

x j
(u j ) =p
i +
x

x


+ + gi

(2)
i j x j xi

(energy equation)

(
eT ) T ( )
pu j
t
+

x j
eT u=
j (
q + )k
x j x j

x j
(3)
Acer Aspire 4720Z
(
ij ui ) + g
(computational fluid +
x j
i
dynamic, CFD)
k standard, renormalization group (RNG),
, ui
realizable
, p ,

 CST-005337
, gi xi standard k - model [4]
x, y, z

t
( ) + x ( ui ) =x + t
+ C 1 Gk

x i k
2.2 k - i j (7)
2
3 C 2
k
standard, RNG realizable
k

= =
C 0.09,C =
1 1.44,C =
2 1.92, k 1.0 = 1.3
(turbulent kinetic

energy equation) RNG k - model [5]

t k
( k ) + ( kui ) = + + Gk - (4)
( ) +

( ui ) =

+ t






t xi x j k xi t x i x j x i
(8)

(
+ C 1P C* 2
k
)
,
k
, C 3 (1 0 )
*
Gk
C 2 C 2 + (9)
1+ 3

Sk (10)


u j
Gk = -uiu j (5)
xi 0 4.38,
= = 0.012,C
= =
1 1.42,C 2 1.68 C = 0.00845

uiu j realizable k - model [6]

Boussinesq [3] (eddy
( ) + x ( ui ) =x

+ t

+ C1S

t i j x i
viscosity) (11)
2
C2
k +
2
k
t = C (6)




(dissipation rate equation) C1 = max 0.43,
+5
(12)

k
=S

(13)
 CST-005337

S 2Sij Sij (14) y + 30

1
C = (15) non-equilibrium wall function [8]
A0 + As kU
law of the wall
pressure gradients
ij
U Sij Sij + ij (16) two layer based

= 2
ij ij ijk k (17) enhanced wall function [9]
2
ij =ij ijk k (18) laminar turbulent
laminar
As = 6 cos (19) turbulent
logarithmic
1 1
= cos
3
( 6W ) (20)
blending function
W=
Sij S jk Ski
(21) enhanced wall function
S

3.
S = Sij Sij (22)


1 u j u i
=
S ij +
2 x i x j


(23) (steady flows)


ij mean rate of rotation tensor

ijk alternating tensor
k - wall
function 3 standard [7], non-
equilibrium [8] enhanced [9]
= =
A0 4.04,C =
1 1.44,C =
2 1.9, k 1.0 = 1.2
(differencing scheme)
SIMPLEC QUICK discretise
wall function Upwind ,

wall function 3
(grid independence)
standard wall function [7]
law of the wall y + 703,977
30 60 y + = u y / u 10-5
 CST-005337

5
1

, 2
, hard disk
3

4
5



1 Ther- ,
mocouple Type K , hard disk
1 12 ,

2 2
1 5
Component Battery CPU RAM Hard disk CD-Rom Computer notebook
Condition 39.29 C 42.15 C 36.37 C 42.11 C 33.71 C 0.34 m X 0.24 m X 0.025 m
o o o o o

1
 CST-005337

2 5

4.


5 4
standard, RNG, realizable

k - wall function 5
3
2 2

k -
5
realizable standard,
enhanced, non-equilibrium wall function 2
2
2

5 0.3 oC
k - realizable non- 6
equilibrium wall function 7

hard disk
RANS 5 2

4-7
k - realizable

non-equilibrium wall function
 CST-005337
3
2
2

5

5
8

, hard disk
5
2
N
experimental data - computational data

i=1
100 %
experimental data
Model Error percentage =
N

standard k : standard wall function 4.2645

standard k : non-equilibrium wall function 4.1344
standard k : enhanced wall function 4.0043
RNG k : standard wall function 4.0358
RNG k : non-equilibrium wall function 3.7756
RNG k : enhanced wall function 3.8459
realizable k : standard wall function 2.9283
realizable k : non-equilibrium wall function 1.0725
realizable k : enhanced wall function 2.9342

3
Average temperature
case
Battery CPU RAM Hard disk
1 37.3304 oC 33.5724 oC 34.0186 o
C 30.0827 oC
2 36.8304 oC 32.3376 oC 33.8622 o
C 29.9347 oC
3 36.5798 oC 33.6462 oC 34.2049 o
C 29.7168 oC
4 36.5604 oC 33.6843 oC 33.8394 o
C 29.4024 oC
5 35.8908 oC 32.6758 oC 33.8105 o
C 29.3251 oC
 CST-005337

40 40
experment

standard k-e model:standard wall function

39 39 standard k-e model:non-equilibrium wall

function
standard k-e model:enhanced wall
Temperature ( 0C )

function
RNG k-e model:standard wall function
38 38
RNG k-e model:non-equilibrium wall
function
RNG k-e model:enhanced wall function

realizable k-e model:standard wall function

37 37

realizable k-e model:enhanced wall

function
realizable k-e model:non-equilibrium wall
functionl
36 36
1 2 3
Locations

39 39

38 38
case 1

case 2
Temperature ( 0C )

37 37

case 3

36 36
case 4

case 5
35 35

34 34
1 2 3
Locations

4 () 5
 CST-005337

35 35

34 34
case 1
Temperature ( 0C )

case 2
33 33
case 3

case 4
32 32

case 5

31 31
4 5 6
Locations

5 () 5

35 35

case 1
34 34
Temperature ( 0C )

case 2

case 3

33 33 case 4

case 5

32 32
7 8 9
Locations

6 () 5
 CST-005337

32 32

31 31
case 1
Temperature ( 0C )

case 2
30 30
case 3

case 4
29 29
case 5

28 28
10 11 12
Locations

7 () 5

8 5
 CST-005337
5.
20
[3] Boussinesq, J., 1877. Theory de Lecoulment
realizable k Tourbillant. Memoires Presentes Par Divers
non-equilibrium wall function Savants Sciences Mathematique at Physiques,
pp. 46-50.

[4] B.E. Launder and D.B. Spalding, 1972. Lectures in
5 Mathematical Models of Turbulence. Academic
Press, London. England.
[5] D. Choudhury, 1993. Introductuon to the
Renormalization Group Method and
Turbulence Modeling. Fluent Inc. Technical
Memorandum TM-107.
6. [6] T.H Shih et al., 1995. A New Eddy-Viscosity
Model for High Reynolds Number Turbulent
Flow. Computers Fluids, Vol. 24, No. 3, pp.
227-238.
[7] Launder, B. E. and Spalding, D. B., 1974. The
7. Numerical Computation of Turbulent Flows.
[1] J.Y. Chang, C.W. YU, and R.L. Webb., 2000. Computer Methods in Applied Mechanics and
Identification of Minimum Air Flow Design Engineering, pp. 269-289.
for a Desktop Computer Using CFD [8] Kim, S. E. and Choudhury, D., 1995. A Near-
th
Modeling. 7 Inter Society Conference on Wall Treatment Using Wall Functions
Thermal and Thermal-Mechanical, Phenomena Sensitized to Pressure Gradient. In ASME
in Electronic System, pp. 330-338. FED 217 ASME.
[2] , , [9] Kader, B., 1993. Temperature and Concentration
, 2549. Profiles in Fully Turbulent Boundary Layers. Int.
FLUENT J. Heat Mass Transfer, pp. 541-1544.

.